MSGNN: Masked Schema based Graph Neural Networks
Summary: Proposes MSGNN, representing HIN neighborhoods via schema instances (minimal complete node contexts) to fuse semantic meta-path advantages with adjacency structure while avoiding manual design. Uses mask-based bi-level self-supervision and a decomposition–reconstruction retrieval; outperforms SOTA (up to +16.08% F1). (summarized by gpt-5-mini on Feb 09 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Hao Liu (Fudan University)
- 2. Qianwen Yang (Fudan University)
- 3. Taoyong Cui (Tsinghua University)
- 4. Wei Wang (Fudan University)
BibTeX Citation
@article{liu_vldb25,
title = {{MSGNN: Masked Schema based Graph Neural Networks}},
author = {Liu, Hao and Yang, Qianwen and Cui, Taoyong and Wang, Wei},
journal = {PVLDB},
series = {{VLDB} '25},
volume = {18},
number = {3},
pages = {571--584},
doi = {10.14778/3712221.3712226},
url = {https://doi.org/10.14778/3712221.3712226},
year = {2025}
}Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 1 of 1 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 65 | Freebase: A Collaboratively Created Graph Database For Structuring Human Knowledge | 2008 | SIGMOD | 0.00038697603 |
Previous
Page 1 / 1
Next